On Sat, Aug 06, 2022 at 11:13:26AM -0700, Nathan Bossart wrote:
> On Fri, Aug 05, 2022 at 03:04:34PM -0700, Andres Freund wrote:
>> But mainly I'd expect to find a difference if the SIMD code were optimized a
>> further on the basis of not needing to return the offset. E.g. by
>> replacing _mm_packs_epi32 with _mm_or_si128, that's cheaper.
>
> I haven't been able to find a significant difference between the two. If
> anything, the _mm_packs_epi* approach actually seems to be slightly faster
> in some cases. For something marginally more concrete, I compared the two
> in perf-top and saw the following for the relevant instructions:
Nevermind, I'm wrong. When compiled with -O2, it uses more than just the
xmm0 and xmm1 registers, and the _mm_or_si128 approach consistently shows a
speedup of slightly more than 5%. Patches attached.
--
Nathan Bossart
Amazon Web Services: https://aws.amazon.com
